This application responds to PAR 10-038 Dissemination and Implementation Research in Health, an initiative focused on the gap between the production of evidence-based interventions and their use in the real world. This problem is exacerbated in organizations like Home Health as most agencies are small, decentralized and geographically dispersed. Few can afford advanced quality improvement. This situation is not unique to Home Health (HH) as much the nation's health care is provided by resource-poor public and not-for-profit agencies. This study tests the effectiveness of a web-based implementation strategy (MHTraining-Net) on the Depression CAREPATH, a HH intervention based on the Collaborative Depression Care model. CAREPATH integrates the model's depression care management functions into routine practice of HH nurses, including: a. Monitoring depression severity; b. Consultation about medication and/or psychotherapy treatment; c. Medication management; d. Education; and e. Patient goal setting around self care, pleasurable activities and social contact. MHTraining-Net employs e-learning modules, webinars, email/phone consultation, and social networking technology to deliver four implementation activities: 1. Infrastructure development, 2. Training, 3. Supervision/performance feedback, and 4. Social learning. The effect of MHTraining-Net on implementing CAREPATH will be compared to an Enhanced Support condition. The study uses an agency-randomized design, with 60 HH agencies per arm. All HH agencies are clients of CareAnyware, Inc, a company that provides web-based clinical management software. The CAREPATH protocol is already programmed into this software and accessible to all agencies. The study will analyze routinely collected data from an estimated 7,800 older patients with clinically significant depression (PHQ-9>15). The aims are to test the effectiveness of MHTraining-Net on: 1. Changing Practice: increasing nurse adherence to the CAREPATH protocol; and 2. Changing Patients' Outcomes: reducing severity of depression. Pilot data suggest that MHTraining-Net is feasible, acceptable to HH agencies, and positively affects both patient care and clinical outcomes. Depression affects over 25% of older HH patients, is mostly undetected or poorly treated, and leads to worse health outcomes and higher healthcare costs. The study takes advantage of technological advances to design an implementation strategy that can reach decentralized, geographically dispersed agencies, and a research strategy that couples the power of an agency-randomized trial with the generalizability of a large, diverse patient population. Given its long reach, MHTraining-Net has the potential of significantly improving care for depression in older HH patients and reducing the public health burden of depression in this population.